The Mu2e Experiment A search for charged lepton flavor violation at FNAL For the Mu2e Collaboration Kevin Lynch Boston University 2010 Fermilab Users' Meeting June 2-3, 2010

2 The roadmap for this talk... Lepton Flavor Violation Primary backgrounds to muon conversion The design of Mu2e

3 The mystery of lepton flavor has been with us for many decades

4 While charged lepton flavor violation has never been observed... However, the predicted SM rates are unobservably small:... today, we know it must occur, even in the Standard Model, through neutrino loop effects.

5 Any observation of cLFV is a direct signal of new physics! The signal kinematics lie in a sea of background from ordinary muon decay. There is no such kinematic limitation for the conversion process. MEG at PSI COMET/PRISM at JPARC Mu2e at FNAL LFV amplitudes typically larger for taus than muons, but production rates dramatically favor muons.

6 Muon conversion arises in many new physics scenarios SUSY Second Higgs Doublet Compositeness Heavy Gauge Bosons Heavy NeutrinosLeptoquarks The discovery of Weak scale SUSY at LHC would imply observable cLFV rates

7 We can parameterize all these models in terms of two effective interactions A dimension 5 dipole interaction A dimension 6 contact interaction And we'll set their relative strength with a dimensionless interpolating factor Active at a common energy scale

8 MEGA SINDRUM II MEG (early) Mu2e Mu2e (Project X) MEG (final) Mu2e probes combinations of these operators deep into the Petascale Mu2e probes well past 10 3 TeV everywhere in this parameter space!

9 The roadmap for this talk... Lepton Flavor Violation Primary backgrounds to muon conversion The design of Mu2e

10 Our muons are not like your muons... Collider Muons Low E Muons Negative muons rapidly stop, capture on atoms, and cascade to the 1s state.

11 Coherent conversion kinematics... a 1-to-2 process producing monochromatic electrons! Starting with a muonic atom in the 1s state......then conversion happens...

12 Proton bunch Pions and muons arrive at target Detector livewindow The two most dangerous backgrounds have very different timing properties. The FNAL accelerator complex produces proton beams with a pulsed structure. Prompt: Radiative Pion Capture with pair production Delayed: Muon Decay-in-Orbit

13 Radiative pion capture can produce electrons near the conversion energy This (and other) prompt backgrounds are eliminated by keeping the DAQ dead until the background rate has fallen below some threshold. Open live window The electron spectrum extends out to nearly m π

14 Electron Energy [MeV] Arbitrary Units Decay-in-Orbit is the major source of delayed background in the live window The free muon decay has a hard cutoff near m μ /2 while the DIO final state has a very long tail which falls very sharply above the Michel peak, to the conversion energy

15 An experimental reality check DIO electrons Conversion electrons Realistic scattering losses smear the distributions to lower energies, while resolution can push events to higher energies.

16 The backgrounds and resolutions constrain the design and reach of the experiment Spectrometer design: Track momentum resolution: O(200keV) Beam transport Out-of-time particle suppression: 10 9 In two years of running This is a factor 10 4 improvement on the current limit!

17 The roadmap for this talk... Lepton Flavor Violation Primary backgrounds to muon conversion The design of Mu2e

18 Where are we going to build this thing? Tevatron Recycler Debuncher Accumulator Booster North You are here The Boomerang Scheme Mu2e Hall

19 Debuncher Cycle 1.7μs Extinction Proton bunch Wait 700ns for prompt backgrounds Debuncher Cycle 1.7μs Proton bunch Pions and muons arrive at target We will utilize unused Booster batches during Noνa running These batches are stacked and bunched in the accumulator, then transferred to the debuncher for slow extraction to Mu2e Mu2e running does not impact the neutrino program! Debuncher Cycle 1.7μs Proton bunch Pions and muons arrive at target Detector livewindow 4x10 12 p

20 What goes in the hall? Unique design features of Mu2e: Pulsed proton source Curved transport solenoid Graded production and detector solenoid fields

21 Muon production solenoid The off-axis 8GeV proton beam delivers 3x10 7 protons per pulse, every 1.7μs The design of the solenoid systems is a collaborative effort among TD, AD, and Physicists: “It's not just a big gizmo for TD to build... we need University people to get involved!”

22 Transport solenoid The curved transport solenoid separates charged particles in the non-bend plane. Collimators in the central straight section reject most wrong sign particles, and can be rotated to change sign for calibration runs.

23 The detector solenoid and stopping target 2.0T 1.0T Graded to reflect electrons toward the tracker 1.0T Uniform to simplify tracking The remnant muon beam and most DIO electrons pass through the central openings, and are caught by a beam stop.

24 The target and detectors 1024 crystals in four vanes with 4MeV resolution at 100MeV The crystal calorimeter provides direct measures of track position, time, and energy to help protect against catastrophic reconstruction errors. The tracker is a novel straw tube in vacuum design. A number of baseline designs are under consideration; all would have or order readout channels. The stopping target is a series of aluminum foils A conversion electron track

25 With this design, the background rates are indeed very small! BackgroundEvents per 2x10 7 s μ DIO0.225 Radiative π capture μ DIF0.072 Scattered beam e Total0.41 In two years of running, we expect fewer than one background event in the signal region. If R μe = , we will see 40 signal events

26 The schedule is technically driven We have received both Stage 1 approval and CD-0. Magnet design and construction remain the schedule drivers.

27 Summer 2010 R&D Tracker Development Straw thickness and composition Winding and sealing Time/charge division and TDC/ADC evaluation Simulation and software Cosmic Ray Veto Efficiency Extinction Monitor Studying SiPMs Calorimeter studies Solenoid design

28 In the Project X era... With no signal All backgrounds must be further suppressed to reach R μe of All detector and transport systems will need upgrades to handle the higher rates If we do see a signal Change target material to determine structure of new physics amplitudes All detector and transport systems will need upgrades to handle the higher rates... beam power will increase fold. Mu2e naturally extends into this era, regardless of the first generation results.

29 Please join us! Boston University Brookhaven National Laboratory City University of New York College of William & Mary Fermi National Accelerator Laboratory Idaho State University Instituto Nazionale di Fisica Nucleare, Lecce Instituto Nazionale di Fisica Nucleare, Pisa Institute for Nuclear Research, Moscow Joint Institute for Nuclear Research, Dubna Laboratori Nazionali di Frascati Los Alamos National Laboratory Muons Inc. Northwestern University Rice University Syracuse University University of California, Berkeley Lawrence Berkeley National Laboratory University of California, Irvine University of Illinois University of Massachusetts, Amherst University of Virginia Contact: Bob Bernstein, or Jim Miller,

30 Backup Slides

31 We normalize the conversions with the captures Events in signal window Acceptance for signal events Directly measured via cascade X-Rays Well known nuclear capture ratio

32 Muon Decay-in-Orbit First, consider the free decay Three body final state

33 Negative muon stopping in matter 1s ground state Electron scattering Atomic capture to high n state Electromagnetic cascade

34 Monitoring the muon stopping rate Tracker Region Calorimeter Region

35 The choice of Al is well matched to the FNAL beam time structure

36 The endpoint energy is material dependent

37 Different materials are sensitive to different operators

38 Nuclear capture is both normalization and a source of inefficiency due to extra hits The lepton-nucleon weak current produces an unstable nucleus that subsequently decays

39 Nuclear capture is both normalization and a source of inefficiency due to extra hits For an 27 Al target, the average capture results in 0.1p, 2n, and 2γ. These result in additional hits in the detectors, potentially leading to inefficiency and pattern recognition errors.

40 Conversion is just one complementary piece of the flavor puzzle MSUGRA while disagreement could point us in the right direction. The ratio of MEG to Mu2e observations Agreement would validate any leading model, LHC and g-2 can tell us where we live in this space

41 Different cLFV processes are driven by different operators Hence, measurements of the different processes tell us about the underlying physics Chirality preserving contact interaction Chirality violating dipole interaction

42 Coherent conversion kinematics The 1-to-2 process produces monochromatic electrons! This energy is well above the vast bulk of the muon decay backgrounds!

43 But the DIO endpoint is the conversion energy DIO electrons Conversion electrons The DIO and conversion energies are the same (up to neutrino masses)